In my last Analyze This post, I alluded to the importance of click to conversion time, that is, the estimated conversion time of the revenue attribution window. I shall devote this article to explaining the importance of the revenue attribution window and some empirical methods to help determine the right window length.

The revenue attribution window refers to the maximum length of time between click and conversion that an advertiser must include when calculating the revenue from a click. Consider two consumers who clicked on an ad for a t-shirt on a given day. One consumer clicked on the ad and bought the shirt on the same day. The other consumer did not convert the same day but bought a shirt from the same website two months later.

Intuitively, we know that the ad was instrumental in the first conversion but perhaps had a very minimal or even no role in the second conversion. If my revenue attribution window was 10 days, I would consider the first purchase but not the second one when I calculated revenue for the keyword for that day, but if my attribution window was 60 days I would include both purchases. Clearly, the attribution window can have a big effect on how profitable a campaign or keyword appears.

So how we set the right attribution window? There are several complex ways to calculate this using methods such as gamma functions, hazard modeling and so on, but I shall present a very simple heuristic using conversion time that will get you started.

The following graph analyzes the click to conversion times of 2500 purchases on a retailers website for a clothing campaign.

Conversions vs. Time

There are some interesting patterns in this chart. First, the conversions are bi-modal. The largest number of conversions takes place in the first hour, after which conversions fall off. The next peak occurs between one and two days. It appears that consumers either buy clothes immediately or wait a day or two before buying, perhaps to comparison shop or ask their significant others for an opinion.

Also of note is that after seven days, the number of conversions drops off. The X axis is not linear as the gap between seven and 10 days and 10 and 15 days is the same. Hence, it appears that after seven days, very few conversions take place.

If I were to plot the cumulative conversions over time, the graph looks as follows:

Cumulative Clothing Conversions

97% of conversions occur in the first 10 days and the remaining 3% take 20 more days to complete. These conversions were recorded with a 30 day attribution window so we are inherently assuming that after 30 days the ad has played no role in the conversion. With this assumption, we can set lower and upper bounds that the revenue attribution window should be between 10 and 20 days.

There are two caveats to this approach. First, we would never know what the upper bound of the window should be. If I waited for 20 years then there might be thousands of additional conversions that will never be recorded when using 30 day window. However, intuitively we assume that an ad had no role to play for a conversion that took place 20 years after the first click. Second, this does not account for the branding effect of an ad. It is probable that ads seen over the years play a role in branding and subsequent conversions but it is hard to measure this effect. Nevertheless, this method when coupled with your business intuition would give you good estimates of the right revenue attribution window.

If you would like to estimate your revenue attribution window, I recommend the following guidelines:

  • Plot the conversion time for product line where keywords behave homogeneously. In my last article, I showed that for some apparel retailers wedding keywords had a very different conversion time as compared to footwear keywords. Hence, I would estimate revenue attribution windows for these campaigns separately.
  • Use a long cookie window to collect your data. In this example we used 30 days. You might want to use a longer window.
  • Plot the cumulative conversions over time for a test period.
  • Calculate the minimum and maximum revenue attribution windows using percentage bounds. For instance, one could use a 95% and 99% cumulative conversion time as the lowest and highest revenue attribution window. In the above example, it would mean that the smallest and longest revenue attribution windows would be seven and 20 days respectively.
  • The click to conversion time metric can reveal the buying behavior of your consumers and can also be used for making important business decisions such as determining the length of your revenue attribution window. Proper use of this metric will give you key insights about your business and will take your online campaigns to the next level of performance.

    Opinions expressed in the article are those of the guest author and not necessarily Search Engine Land.

    Related Topics: Channel: Analytics | Search & Analytics

Sponsored


About The Author: is Director, Business Analytics at Adobe. He leads a global team that manages the performance of over $2 BN dollars of ad spend on search, social and display media at Adobe.

Connect with the author via: Email | Twitter | Google+ | LinkedIn



SearchCap:

Get all the top search stories emailed daily!  

Share

Other ways to share:

Read before commenting! We welcome constructive comments and allow any that meet our common sense criteria. This means being respectful and polite to others. It means providing helpful information that contributes to a story or discussion. It means leaving links only that substantially add further to a discussion. Comments using foul language, being disrespectful to others or otherwise violating what we believe are common sense standards of discussion will be deleted. Comments may also be removed if they are posted from anonymous accounts. You can read more about our comments policy here.
  • http://www.onlinematters.com Arthur Coleman

    Excellent article. I love how you took the analysis to a first integral in order to show how the bimodal distribution turns into a smooth logarithmic function when cumulative sales are considered. As a retailer, you could also use this same analysis to look at the impact of merchandising (e.g. BOGO, 20% off one day sale) and then can more accurately predict sales from merchandising of various types.

    Now the hard question: what if my first visit is from a click on Google and my second from a click on Bing? How would you go about dealing with the data to get the same accurate analysis? Now you have two cookies and two apparently different people. We have figured this out through our internal analytics, but not sure we have still got it right. Tough problem to solve.

  • zwelling

    I would add a third point to your list of caveats: you are assuming your retailer has perfect information. We’ve learned, in practice, that even with the longer cookie window you advise, information gets lost to cookie deletion and cross-browser research/shopping. Our customers see more conversions later in the attribution window than your data would suggest. One quick analysis we ran indicated that 14% of revenue delivered to a client of ours after the first day would’ve been misallocated to incorrect (more recent) events were it not for our proprietary tracking system.

    Jeff Zwelling CEO Convertro.com

 

Get Our News, Everywhere!

Daily Email:

Follow Search Engine Land on Twitter @sengineland Like Search Engine Land on Facebook Follow Search Engine Land on Google+ Get the Search Engine Land Feed Connect with Search Engine Land on LinkedIn Check out our Tumblr! See us on Pinterest

 
 

Click to watch SMX conference video

Join us at one of our SMX or MarTech events:

United States

Europe

Australia & China

Learn more about: SMX | MarTech


Free Daily Search News Recap!

SearchCap is a once-per-day newsletter update - sign up below and get the news delivered to you!

 


 

Search Engine Land Periodic Table of SEO Success Factors

Get Your Copy
Read The Full SEO Guide